AMD Zen 3 (5000 Series), rumored 17% IPC gain.

[4K8KW10]

I think it has more to do with the heat *density*. AMD's power consumption is low, but it's crammed into a smaller area.

AMD's power consumption is not lower if you look at the Ryzen 5 vs Core i5 benchmarks - https://www.techpowerup.com/review/intel-core-i9-10900k/18.html

It's hilarious how after so many years, the Intel CPUs are still extremely competitive and using a 5-year old process.

Not good for AMD.

 

[Vogon]

Yes, I will need to adjust my expectations, coming from an i5 that games at mid 40s and under cpu stress tests peaks at 72

 

[jigger]

AMD's power consumption is not lower if you look at the Ryzen 5 vs Core i5 benchmarks - https://www.techpowerup.com/review/intel-core-i9-10900k/18.html

It's hilarious how after so many years, the Intel CPUs are still extremely competitive and using a 5-year old process.

Not good for AMD.

Have you read that review? The 3000 chips are few percent slower in games. Faster at everything else and pull about half the power...

 

[DeathStorm]

AMD's power consumption is not lower if you look at the Ryzen 5 vs Core i5 benchmarks - https://www.techpowerup.com/review/intel-core-i9-10900k/18.html

It's hilarious how after so many years, the Intel CPUs are still extremely competitive and using a 5-year old process.

Not good for AMD.

Oh please, so when you're not stressing the CPU the Intel comes out decently enough but otherwise is quite poor in comparison even before we factor in performance gains. Conclusion? Not good for AMD.

AMD about to release some new CPUs? Just think of how much money they'll lose on packaging, not good for AMD!

 

[jigger]

Oh please, so when you're not stressing the CPU the Intel comes out decently enough but otherwise is quite poor in comparison even before we factor in performance gains. Conclusion? Not good for AMD.

AMD about to release some new CPUs? Just think of how much money they'll lose on packaging, not good for AMD!

I think he just posts nonsense and hopes nobody will notice. The chap might not be wired properly or is just looking to flame bait.

 

[4K8KW10]

so when you're not stressing the CPU the Intel comes out decently enough but otherwise is quite poor in comparison even before we factor in performance gains. Conclusion? Not good for AMD.

AMD about to release some new CPUs? Just think of how much money they'll lose on packaging, not good for AMD!

Not good for AMD.
Performance gains? lol There are no performance gains, actually AMD lags behind in the performance part.

Relative performance https://www.techpowerup.com/review/intel-core-i9-10900k/21.html
CPU Tests:

 

[jigger]

Not good for AMD.
Performance gains? lol There are no performance gains, actually AMD lags behind in the performance part.

Relative performance https://www.techpowerup.com/review/intel-core-i9-10900k/21.html
CPU Tests:

Yeah, unfortunately not so much in the real world.

 

[RavenXXX2]

 

[jigger]

Oh I’m stealing that.

 

[Twinz]

AMD's power consumption is not lower if you look at the Ryzen 5 vs Core i5 benchmarks - https://www.techpowerup.com/review/intel-core-i9-10900k/18.html

It's hilarious how after so many years, the Intel CPUs are still extremely competitive and using a 5-year old process.

Not good for AMD.

???

 

[Twinz]

Not good for AMD.
Performance gains? lol There are no performance gains, actually AMD lags behind in the performance part.

Relative performance https://www.techpowerup.com/review/intel-core-i9-10900k/21.html
CPU Tests:

I don't see the 5000 series tested on this chart.

If what AMD has said pans out. Intel has lost the last little sliver of ground they had to stand on.
If independent test verify that Zen 3 does indeed out perform Intel in 1080p gaming now, Intel will have to try and position themselves as the "value" offering. (Assuming AMD keep their prices high.)

Maybe intel can still find the random application where they beat AMD and tout those. lol

 

[Mr Oblong]

Not good for AMD.
Performance gains? lol There are no performance gains, actually AMD lags behind in the performance part.

Relative performance https://www.techpowerup.com/review/intel-core-i9-10900k/21.html
CPU Tests:

I'm not sure you have looked into this properly. There are numerous examples of the Intel chips running hot and thirsty, it's inevitable on 14nm pushing 5Ghz.

 

[Shaolin Chicken]

As it happens you are sort of correct. AMD 2 does run hot, but only when you are hammering the cpu in Prime 95, or OCCT with AVX. Otherwise, even running Realbench, my 3950x rarely get's above 65c. Is my rig "silent" ? no................neither is anyone else's regardless of what they say.
Don't worry about Ryzen temps, as long as you have half decent cooling, it will be fine.

What cooling are you using for your 3950x please?

 

[gpuerrilla]

Wow, new lows for 4k8k..

 

[Steampunk]

Intel have literally shaved the heatspreader down to the minimum so that they can pump more watts though the chip in order to try and keep 14+++++++++++++ nm chips in touch with the 3000 series and not have it melt through the motherboards. And that is on the single metric of single thread gaming at low resolutions. The 5000 series will overtake them in that last little bastion of (supposed) performance benefit. Intel has taken their technology as far as it can go, and this is the end of the road - they need something dramatically new, and they've shown nothing like that so far.

 

[Journey]

Intel have literally shaved the heatspreader down to the minimum

They actually made it thicker, the die/package was made thinner to allow better thermal efficiency.

 

[sideways14a]

What absolute tripe some of the posters on this forum produce.

Have you ever thought about selling your output to farmers, a good load of spreading does wonders for crops and this place has large amounts crap.

 

[Steampunk]

They actually made it thicker, the die/package was made thinner to allow better thermal efficiency.

You are correct, and I stand corrected. Same point though! Intel are stretched to the max on what they can do with their current tech, and are having to clutch at straws like shaving the die. I actually think it's a clever solution to a problem of their own making, but it doesn't fix the basic problem that Intel are at the end of the road with their endless refinement of 14nm. There's no more low or high hanging fruit left, they need something new. AMD were in the same situation and managed to sidestep the problem with chiplets, but Intel are dragging behind.

 

[4K8KW10]

You are correct, and I stand corrected. Same point though! Intel are stretched to the max on what they can do with their current tech, and are having to clutch at straws like shaving the die. I actually think it's a clever solution to a problem of their own making, but it doesn't fix the basic problem that Intel are at the end of the road with their endless refinement of 14nm. There's no more low or high hanging fruit left, they need something new. AMD were in the same situation and managed to sidestep the problem with chiplets, but Intel are dragging behind.

I think Intel wins in other areas - like chipsets / motherboards power consumption, optimising their 14nm++ CPUs to be very power efficient playing with fine tuning of the frequencies, better binning, etc, all the things that AMD is not so good with.

 

[Steampunk]

I think Intel wins in other areas - like chipsets / motherboards power consumption, optimising their 14nm++ CPUs to be very power efficient playing with fine tuning of the frequencies, better binning, etc, all the things that AMD is not so good with.

That's rather like saying that Intel have really, really perfected the solid rubber tyre, so it's much better than the modern air-filled steel belted things that AMD want you to put on your wheels instead.